Bastard Fault A600 Composite Video
 

Only the brave should read on!

A600 rev 2D, perfect picture through scandoubler on VGA monitor. Over coloured picture on composite with picture shifted up and left on screen. No other faults. Crystal checked and definitely PAL

CXA1145 replaced, all caps replaced, resistor networks on RGB in to CXA1145 measured out ok from Q211, Q212, Q213. All voltages the same between a working rev 1.5 on CXA1145. HCT244 both working, Denise resoldered.

All circuitry leading to composite connector ok, cannot test for RF out, no TV and video remote with my folks. VVRef ok, VIO / VID ok.

Cheers.

did you check composite input on the TV itself ?

It's being fed into a DVT-1000 video acquisition card running on a PC and the exact same input is being used for the rev 1.5, which is using the exact same cable.

I don't actually own a television, haven't had one for the last 10 years or so, odd I know for an electronics tech.

Since you can't test RF, have you tried removing C236. This should disconnect it, in case an RF fault is messing up composite somehow.

I had considered that too and removed the entire RF module off the board, needed to check for more capacitor leakage anyway, no difference.

A few more clues perhaps, it initially had a rolling and tearing screen on my test 1084 which I put down to the 1084 having finally died, I only use it to test Amiga units for composite.

What kind of scan doubler are you using? An IndiECS will work even if the RGB output is faulty. What I mean to ask is, can you test the actual RGB port?

If it is OK but but the composite output has disturbed colour and position, maybe there is a fault in the delay line/BPF, external to the CXA chip. I don't really understand this part of the circuit but I think it is important for merging the colour and sync signals into one.

MV1200 Scandoubler.

From my dark days in dealing with video signals, I know it's 64uS per frame which includes colour burst, video signals ( RGB ) sync and back porch. All the analog RGB goes to the video port on the Amiga and I probably concur with the delay line / BPF stage but like you my understanding is VERY limited. I did change the BPF CAN with no difference. One other issue is that the 2D has differences from the rev 1.5 and I only have the circuit for the 1.5. If anyone does have a 2D circuit, no matter how cruddy the quality, could they email it to me please?

zero242@internode.on.net

Cheers.

You can find it here in very good quality:
http://www.amigawiki.org/doku.php?id...ice:schematics

@Loedown,

Got any pictures of the distorted video output?
Is the colour information correct?

Ian

@Demolition

Those schematics are awesome, don't know who did the redraw for those but they should have gotten a medal.

@Ian

Pictures included of both Composite and VGA via MV1200

@FOL

Pictures of the video sections before I repaired them

Nice leakage, ;).
Did'nt know I replied in the thread, lol.

You checked the through holes? also the resistors 2x 75ohm and ferrite bead.

Holy leakage, Batman!

Looks like my old A600 mobo :laughing

Good luck in repairing this, matey. When it comes to A600 composite repairs I'm always interested :D

You didn't reply here FOL but in the thread where I put the fear of God into people about their A600 units.

I am very thorough in checking for problems in boards, the logical place for a fault would be between the CXA1145 and the Composite Out port. I think Ian will have an answer as his knowledge of video is far superior to mine.

I had to remove the capacitors and in both cases the 4k7 resistor as well just to clean up the leakage. The 4k7 resistor needed to have the ends wire brushed to get the corrosion off and in several places the solder resist needed scraping back, cleaning and applying new varnish. My client only wants two of the three repaired which at least gives me one A600 as a parts unit. If there were just more demand for these sorts of repairs then I could use my creative mind to consider better repair techniques and cleaning possibilities.

In my absence I have designed and built a custom reflow oven from a toaster oven and designed a very extensive cable tester, the boards should be here this week for the prototype and then the cases need to be manufactured, I already have the basic fascia.

First two pics are the oven and related, the latter the cable tester fascia.

Really nice! I love the Oven! I'm also interested in the cable tester, don't forget to take a lot of hardware pr0n pics! :bowdown

Hi,

Looking at the pictures, of both the screen and the motherboard I can see the following:


The CXA1145 is working, it provides the RGB outputs for the MVA1200.[/LIST]

• The same CXA1145 used for the composite output, has luminance/chrominance issues.

• The big 4R7 resistor, is not used by the CXA1145, it is used for the RF modulator so we can ignore it.

• The VVREF supply to part of the CXA1145 is unknown quality due to PCB damage.

Therefore the most likely cause is related to the composite video encoding. The CXA1145 uses an external Bandpass filter (Z221) and delay line (Z222) which have been changed. There are some support components, R221-R225 and C221-C222 that should be checked and if necessary replaced.

I suspect the issues are due to the chrominance carrier of the composite video signal being incorrect, possibly with a degraded luminance component. Pins 15-18 of the CXA1145 have LUMA/CHROMA signals there, with the aforementioned components.

Do you have an oscilloscope for any measurements?

Bye,

Ian

Thanks for all that Ian, yes I do have an oscilloscope. The band pass filter was changed with another board so that removes that from the equation. I didn't change the delay line parts but will look at those components now that I have the updated 2D circuitry. I was leading towards all that but I was hesitant as it's a pain to remove and replace those cans / filter networks / whatever they are.

To everyone that's helped with this repair I am grateful but I am bowing out of it. Checked all the connections between the CXA1145 and it's support circuitry and everything is fine. All the voltages on the CXA1145 of the 2D board were almost identical to the voltages on the R1.5 and there's just nothing I can see beyond what I have done and it's banked up several repairs I need to get done.

I reiterate my warning about A600 boards, don't be complacent, get them recapped and keep an eye on the state of them because the corrosion on these as could be seen was as bad as an A4000 with the original battery. I didn't take any photos of the two caps on the far side of the first board but they were really bad too, 22uF